AI-102 Practice Test 32 — 15 Questions

Question 1

You are a data scientist at a healthcare startup. You have deployed a custom object detection model using Azure Custom Vision to detect tumors in MRI scans. The model was trained on 10,000 labeled scans from a single hospital. After deployment, the model performs well on scans from that hospital but poorly on scans from a different hospital with a different MRI machine. The new hospital's scans have slightly different contrast and resolution. The model's precision drops from 0.92 to 0.65, and recall drops from 0.88 to 0.50. You have access to 500 labeled scans from the new hospital. You need to improve the model's performance on the new hospital's data as quickly as possible with minimal effort. What should you do?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Use the existing model as a starting point and retrain it with the 500 labeled scans from the new hospital.. Azure Custom Vision supports transfer learning, allowing you to take an existing trained model and retrain it with new labeled data. By using the 500 labeled scans from the new hospital as a training set, you can fine-tune the model to adapt to the different contrast and resolution characteristics without starting from scratch. This approach is the fastest and requires minimal effort, leveraging the previously learned features while incorporating domain-specific adjustments.

Answer

Collect more labeled scans from the new hospital and train a new model from scratch.

Answer

Create a new Custom Vision project and train only on the 500 new scans.

Answer

Apply image preprocessing to normalize the new hospital's scans to match the old hospital's style, then use the existing model.

Question 2

You are a generative AI engineer at a financial services company. The company uses Azure OpenAI Service to generate investment summaries. You have deployed a GPT-4 model with a content filter set to 'Low' for hate speech. The model frequently generates summaries that include biased language against certain demographics. You need to reduce biased outputs while maintaining the ability to generate detailed financial analysis. You cannot afford to retrain the model. You have the following options: A) Change the content filter severity to 'High' for all categories, B) Add a system message instructing the model to avoid bias and provide examples of unbiased summaries in the prompt, C) Use the Azure AI Language service to detect bias in the output and regenerate if bias is found, D) Deploy a different model like GPT-3.5 which has less bias. Which course of action should you take?

Accepted Answer

Add a system message instructing the model to avoid bias and provide examples of unbiased summaries in the prompt.. Adding a system message and examples in the prompt (few-shot learning) can effectively reduce biased outputs without retraining. Option A is wrong because increasing content filter severity may block legitimate financial analysis content. Option C is wrong because detecting bias after generation and regenerating is inefficient and may still produce biased output. Option D is wrong because GPT-3.5 may also exhibit bias and may not provide the same quality of financial analysis.

Answer

Use the Azure AI Language service to detect bias in the output and regenerate if bias is found.

Answer

Change the content filter severity to 'High' for all categories.

Answer

Deploy a different model like GPT-3.5 which has less bias.

Question 3

A company is building an agent that uses Azure OpenAI to answer questions from a large document library. The agent must use a Retrieval Augmented Generation (RAG) pattern. Which TWO actions should the team take to implement RAG effectively?

Accepted Answer

Index the documents into a vector database like Azure Cognitive Search.. Indexing documents into a vector database like Azure Cognitive Search enables efficient similarity search over embeddings, which is the retrieval foundation of RAG. This allows the system to quickly find the most relevant document chunks based on semantic meaning, rather than relying on the model to memorize or be fine-tuned on the entire library.

Answer

Ensure the model is large enough to memorize the entire document library.

Answer

Fine-tune the Azure OpenAI model on the document library.

Answer

Train a custom language model from scratch.

Question 4

You see the exhibit representing an Azure Bot resource configuration. The bot is not responding to user messages. What should you verify first?

Accepted Answer

Check that the endpoint URL is correct and the web app is running. The most common reason a bot fails to respond to user messages is that the endpoint URL configured in the Azure Bot resource does not match the actual URL of the running web app, or the web app itself is stopped or unresponsive. Without a correct and reachable endpoint, the Bot Framework Service cannot forward messages to the bot's message handler, causing a complete communication breakdown.

Answer

Ensure that LUIS app IDs are provided

Answer

Verify that the Application Insights key is valid

Answer

Confirm that the msaAppId is a valid GUID

Question 5

You deploy the ARM template shown in the exhibit. After deployment, you need to allow access to the Language service from your on-premises application. What should you do?

Accepted Answer

Add an IP rule with your on-premises public IP address.. The ARM template deploys an Azure Cognitive Services Language service with a network ACL that defaults to denying all traffic (defaultAction: Deny). To allow your on-premises application to access the service, you must add an IP rule that permits traffic from your on-premises public IP address. This is because the network ACL evaluates IP rules before the default action, and adding a rule with your public IP overrides the default deny for that specific source.

Answer

Remove the customSubDomainName property.

Answer

Set the defaultAction to Allow.

Answer

Change the SKU to F0 to allow public access.

Question 6

You are a lead AI engineer for a global retail company. The company is building an AI-powered customer support chatbot using Microsoft Foundry. The chatbot must answer product questions, process returns, and escalate to human agents when needed. The solution uses Azure AI Language for intent recognition and Azure AI Bot Service for bot orchestration. During testing, the chatbot fails to understand customer queries about return policies. The intents 'ProductInquiry' and 'ReturnRequest' are defined, but the model often confuses them. You need to improve intent classification accuracy. The development team has already collected 500 sample utterances for each intent. You have a budget to collect additional data. What should you do?

Accepted Answer

Collect additional utterances that are similar to the confusing ones and retrain. Collecting additional utterances that are similar to the confusing ones directly addresses the ambiguity between the 'ProductInquiry' and 'ReturnRequest' intents. By providing more representative examples of edge cases where the intents overlap, the Azure AI Language model can better learn the subtle linguistic patterns that distinguish them, thereby improving classification accuracy without requiring a complete retraining from scratch.

Answer

Use Azure AI Translator to translate utterances into multiple languages

Answer

Increase the confidence threshold in the bot configuration

Answer

Create a new custom language project and retrain from scratch

Question 7

Which ONE Azure AI service can be used to perform sentiment analysis on text?

Accepted Answer

Azure AI Language. Azure AI Language provides built-in sentiment analysis capabilities as part of its natural language processing (NLP) features, allowing you to assess the sentiment (positive, negative, neutral, or mixed) of text at the document and sentence level. Azure AI Translator does not include sentiment analysis; it only provides text translation. Therefore, only Azure AI Language is correct.

Answer

Azure AI Search

Answer

Azure AI Bot Service

Answer

Azure AI Translator

Answer

Azure AI Content Safety

Question 8

You are developing a solution that uses Azure AI Video Indexer to analyze surveillance videos for suspicious activity. The solution must generate alerts when a person is detected in a restricted area. Which feature should you use?

Accepted Answer

Azure AI Video Indexer people detection and tracking. Azure AI Video Indexer's people detection and tracking feature is specifically designed to detect and track individuals across video frames, making it ideal for generating alerts when a person enters a restricted area. This feature provides bounding boxes, timestamps, and tracking IDs that enable real-time monitoring and alerting based on spatial rules.

Answer

Azure AI Video Indexer sentiment analysis

Answer

Azure AI Face identify API

Answer

Object detection in Azure AI Vision

Question 9

An organization uses Azure AI Search to power an internal knowledge base. They notice that search results are returning irrelevant documents. The index includes a 'content' field with full text and a 'tags' field with metadata. Users often search for specific terms that appear in the 'tags' field. How should you configure the search index to improve relevance?

Accepted Answer

Configure a scoring profile with a higher weight for the 'tags' field.. Configuring a scoring profile with a higher weight for the 'tags' field increases the relevance score of documents where search terms match the tags, thereby prioritizing those results. Option A (freshness-based scoring) would favor newer documents but does not address matching on tags. Option C sets the 'tags' field to use the 'keyword' analyzer, which changes tokenization but does not adjust field weighting. Option D enables semantic search on the 'content' field, which enhances understanding of natural language queries but does not specifically boost the weight of the tags field.

Answer

Add a custom scoring profile based on freshness.

Answer

Set the 'tags' field to use the 'keyword' analyzer.

Answer

Enable semantic search on the 'content' field.

Question 10

Your company has a large repository of scanned invoices in PDF format. You need to extract invoice number, date, total amount, and vendor name from these PDFs. Which Azure AI service should you use?

Accepted Answer

Azure AI Document Intelligence. Azure AI Document Intelligence (formerly Form Recognizer) is purpose-built for extracting structured data (fields like invoice number, date, total amount, vendor name) from scanned documents such as invoices and receipts. It uses prebuilt models trained specifically on invoice layouts and supports both OCR and key-value pair extraction from PDFs.

Answer

Azure AI Language

Answer

Azure AI Search

Answer

Azure AI Vision

Question 11

You are implementing a generative AI solution using Azure OpenAI. You need to ensure that the model's outputs do not contain certain inappropriate words or phrases. Which feature should you configure?

Accepted Answer

Content filters. Content filters in Azure OpenAI are specifically designed to detect and block inappropriate words or phrases in both prompts and completions. They operate at the service level, applying configurable severity thresholds for categories like hate, violence, sexual content, and self-harm, ensuring model outputs adhere to policy without requiring prompt engineering or data modifications.

Answer

System message instructions

Answer

Grounding with your data

Answer

Max tokens limit

Question 12

A healthcare organization is deploying a solution using Azure AI Language to extract medical entities from clinical notes. The solution must comply with HIPAA and support the following requirements: extract medication names, dosages, and frequencies; identify patient conditions; and recognize negated terms (e.g., 'no sign of infection'). Which THREE Azure AI Language features should the organization use?

Accepted Answer

Prebuilt NER for Healthcare. Prebuilt NER for Healthcare is specifically designed to extract medical entities such as medication names, dosages, frequencies, and patient conditions from unstructured clinical text. It is a HIPAA-eligible Azure service that provides domain-specific entity categories, making it the correct choice for the healthcare use case described.

Answer

PII detection

Answer

Prebuilt NER for Finance

Question 13

Your company runs a global e-commerce platform. You are building a chatbot using Azure AI Language's conversational language understanding (CLU) to handle customer requests in multiple languages. The bot must support English, German, and Japanese. You have labeled training data in English only. The deadline is tight, and you want to minimize manual labeling. You also need to ensure that the bot can gracefully handle unsupported languages (e.g., French) by directing the user to a human agent. You have access to Azure AI Translator. Which approach should you take?

Accepted Answer

Use a single CLU project with multilingual option enabled, train on English data only. Configure the bot to detect the language of user input; if it is English, German, or Japanese, route to CLU; otherwise, route to a human agent.. Azure AI Language's CLU supports a multilingual option that allows a single project to handle multiple languages without requiring translated training data. By enabling this option and training on English data only, the model can generalize to German and Japanese due to shared multilingual embeddings. The bot can then detect the user's language and route unsupported languages like French to a human agent, minimizing manual labeling while meeting the deadline.

Answer

Use a single CLU project with English data only. Translate the English training data into German and Japanese using Azure AI Translator, then train a single multilingual model by including the translated data.

Answer

Use a single CLU project with English data only. Before calling CLU, translate non-English user input to English using Azure AI Translator. For unsupported languages, detect language and route to human agent.

Answer

Build separate CLU projects for English, German, and Japanese. Label training data in each language by translating the English data using Azure AI Translator.

Question 14

You are deploying an Azure AI Search solution for a global e-commerce platform. The index must support real-time updates from a product catalog stored in Azure Cosmos DB, with a maximum indexing latency of 10 seconds. The solution must also handle 5000 queries per second (QPS) during peak hours. What should you configure?

Accepted Answer

Use Standard S3 tier with 12 replicas and enable incremental indexing.. The Standard S3 tier supports high query volumes (up to 5000 QPS) with 12 replicas, and incremental indexing enables near-real-time updates from Cosmos DB by processing only changed documents, keeping latency under 10 seconds. The combination of sufficient replicas for throughput and incremental indexing for low-latency indexing meets both requirements.

Answer

Use the Storage Optimized L2 tier with 6 partitions and 4 replicas.

Answer

Use the Free tier and enable change tracking on Cosmos DB.

Answer

Use Standard S2 tier with 3 replicas and indexer scheduling every 5 minutes.

Question 15

Drag and drop the steps to configure a multi-region disaster recovery for Azure Cognitive Services into the correct order.

Accepted Answer

Create resources in both regions, then enable geo-replication, then configure failover routing, then set up monitoring, then test failover. Start by creating resources in both regions, then set up replication, configure failover routing, monitor health, and test.

Answer

Create resources, then configure failover routing, then enable geo-replication, then set up monitoring, then test failover

Answer

Create resources, then enable geo-replication, then set up monitoring, then configure failover routing, then test failover

Answer

Create resources, then enable geo-replication, then configure failover routing, then test failover, then set up monitoring